Ask HN: What's Wrong with WordPress?

Despite being a dev of 20+ years, I use Wordpress as a non-technical user, hosted on a paid for plan on Wordpress.com. I never deploy, never have to upgrade, never have to troubleshoot issues. Whatever monolithic and legacy complexity is part of Wordpress I don't see it.

I use it regularly - 3 or more posts a week - and it's as fast as I could hope for. So easy to create and edit posts. So many plugins for just about everything I want to do. And professional templates - with support - are so cheap I'm almost embarrassed to be paying for it.

I think a lot of people see Wordpress as it was back in the early 2000s, when amateur bloggers were liking each others posts and getting comments was the goal. And fighting spam was a daily chore.

I see it as a simple publishing platform where I can focus on writing and not on infrastructure. I really don't care what's under the covers. If your better solution is any more complicated than me signing up, I'm not interested.

Are people genuinely building new things with Wordpress these days?

Years ago there was a surge of people putting out blogs and websites like that which brought demand for plug-ins, etc. Long form text content isn't the trend now, its podcasts and vlogs, etc. In some ways, amateur content creators just need a smartphone.

I think if someone was going to build a platform for nearly anyone to publish content it would need to have easy plugins to do things with video and audio and not be focused just on text. Allowing people to build plug-ins and themes and install them is a key feature.

I think the technology choice that would make this easiest these days is JavaScript (even tho I'm not a user). The numbers of people out there today with deep JS skills are similar to when there were a lot of PHP developers building WordPress plugins. It allows for front and back end development which could be an advantage to a plugin based tool like WordPress is.

I wish it were a page builder like Wix with different drop-in modules (here's a blog, here's an events calendar, here's a marketing page, here's an ecommerce section, here's a photo gallery), as opposed to a blog engine first and everything else shoehorned in.

I also wish it weren't PHP or require such a heavy backend stack. Having something like the above but serverless and portable would be pretty cool. Open source internals, paid templates, pick any cloud to host on, 1-click deploy, 10% of the hosting fee goes back to open source foundation, then any editor can drag and drop and rearrange modules as needed. Oh, and no for-profit branch run by the same people competing against the other vendors...

IMHO, writing themes for WordPress seems very convoluted. I much prefer how themes are handled in Hugo, which IIRC is based on Go’s standard templating engine. I’ve tried reading up on how to make WordPress themes, but instead of actually writing a theme I ran in the other direction and went with Hugo instead.

To me, this feels like an anti-feature of WordPress. In a sane CMS, getting a basic theme up and running – from scratch – should be easy.

Rebuilding it today and how I would build it then if I knew what I know now are two different questions. There is also the question of what went wrong. I don't like the block editor introduced in 2018.

If I were building it today, I would write it in golang and I would offer the ability to export static sites. I would make comments an optional or external module since spam has largely ruined comments sections for anyone who doesn't get many legitimate comments. Why golang? Because it typically has the easiest deploy story, IMO.

If I were writing it then, I would have written it in Python and chosen PostgreSQL as the database. Of course, back then that is what I would have chosen as well.

What I remember from the time I used Wordpress (about 10 years ago) was that everybody was trying to spam links on your site, that could include spam comments but would also involve people taking advantage of any security loophole they could find. If you kept patched you might be alright but if you walk away for a few months you might find spam links in places you didn’t even know your site had.

I remember people always talking about installing caching plugins which made me wonder why caching isn't built in.

I would probably add more features to make it usable as a CMS, rather than just a blogging platform. At the moment, custom fields are basically unusable without a plugin like Advanced Custom Fields, and the same is true of custom post types and formats, the user login system, etc. Lots of things that could make it a better general CMS are entirely left to plugins whereas they probably shouldn't be.

I also don't get all the focus on full-site editing and blocks and customisation and Gutenberg either, since it feels like these features are only really going to be usable if either you're running your site as a solo venture or you're working in some sort of media environment where fancy one-off article styles can be the norm (like some New York Times longform pieces).

Anything other than that feels like a place where the modern style of editor is just a recipe for disaster. A large corporation doesn't want its editors to go wild with blocks and widgets in their posts and pages. An agency doesn't want the client to play designer and go crazy with the theme customisation. They're features meant for the Squarespace/Wix market, and not ones that necessarily fit something like WordPress.

But eh, it feels like there's too much focus on trying to compete with page builders and hosted site building tools, and not enough focus on fleshing out the features already included there, or providing a solid, easily extendable editing experience for existing WordPress site owners.

It's older, slower, and clunkier than more modern alternatives. It's a monolith in the age of microservices